Robots are usually associated with handling repetitive tasks - either in high volume production or flexible handling systems for frequent changes. In the food packaging industry, robots generally fall into three main arenas: pick and place applications, feed placement and palletising.
Labour savings, reduced sickness benefits, the overcoming of labour shortages, better product quality and improved working conditions are the most obvious benefits associated with robotics systems. Less obvious are savings such as a reduction in cafeteria facilities, staff recruitment and training costs, National Insurance contributions and even the number of parking spaces required.
It is important to optimise the line by achieving the highest line and robot performance, the minimum number of robots, the highest efficiency, low robot load, high reliability and high redundancy. But while these objectives are easily defined, they are less easy to fulfil because of boundary conditions and other external factors in most packaging processes.
The main issue relates to product and tray flow. Variations and inconsistencies in the product’s rate, missing or damaged items and starting/stopping of the line all have a major impact on its efficiency, irrespective of the capabilities of the robots. Other limiting factors include the specific placing order of different product types, the introduction of cushions between trays, the grippers required to handle delicate products, and the potential for malfunction of ancillary equipment.
The number of products in the flow should be equal to the required number of products to fill all the trays. On most lines, there is always the possibility of either having too few products to fill the trays or an uneven product distribution. This prevents the trays from moving constantly and consistently to the end of the line and causes frequent line stops and starts.
It is important to model the actual scenario in relation to the real time of the robot operations. For example, a line may be designed to deliver a maximum product feed of 300 pieces per minute, but only intermittently peaks at this figure and, when measured in 60-second intervals, actually has a lower rate. However, shortening the interval to 20 seconds demonstrates that the peak flow is achieved about once every minute.
When the variation in product rate is investigated at time intervals of two seconds, it can be seen that the peaks are above 300 pieces per minute at any time. It is this variation that makes the configuration of the robot so important. For the robots to operate at peak efficiency, it is essential they work at a realistically calculated nominal rate.
The line can also suffer from high product overflow, frequent stops of the tray belt and varying performance of individual robots. However, with careful line optimisation it is possible to run the same line consistently, with no product overflow, no tray belt stoppages and an equal performance of all the robots.
Such improvements in efficiency are achieved by adjusting the product belt speed commensurate with the robot performance. That said, efficiency is the name of the game and the number of packed products should always be maximised.
Apart from the product feed, line configuration and belt speeds, the positioning of the robots is critical to attain the highest efficiency. Commonly, robots are configured to pick from the shortest distance possible at their respective station position. This is all well and good at the start of the line, but by the time the fifth or sixth robot is reached, the product can be entirely centred within the belt. This can lead to catastrophic collisions at the end of the line, as the last two robots have to vie for the final picking of the products.
With optimal robot layout, product placement, feed consistency and belt speed, it is perfectly feasible to have the robots picking consistently across the belt without the chance of collision and with minimal overflow. By defining the work area of each robot, there is no need for the reach of any one to encroach into the picking area of another.
While many will accept the cost justification of robotic lines, it is often believed that they are complex to integrate, and hence, have associated hidden costs. However, the advances and innovation in robot technology have come a long way since robots were first used on production lines. Many robot cells are modular and can be pre-built before installation. All the end-user needs to do is put them in position, bolt them together and install the robot. With this ‘plug and play’ concept, installation can be achieved in just one day.
Depending on the design of the robot cell, foundations are often unnecessary. Robots create a large dynamic force that wants to pull them out of the ground, so they need to be securely anchored. But robots can be built, for example, on steel plates in a sandwich board construction that distributes the forces evenly across the surface and massively reduces the dynamic force going into the floor. Delta robots, such as ABB’s IRB340 FlexPicker, are top mounted, and as such consume minimal floor space and are easily integrated. There are many design concepts that make robots easy to integrate, and which can even eliminate some costs altogether.
Once the case for the robot has been analysed, it is important to address the economics of the process. Measured against human costs, this may be justified, but once line efficiency has been considered, payback may come not just from reduced costs, but also from increased productivity, efficiency and quality.
To achieve this, it is essential to evaluate the packaging line, its layout, parameters, throughput and the potential robot configuration.
Frank-Peter Kirgis is with ABB Robotics
Frank-Peter Kirgis’ first article, which examined the use of vision systems in robotic packaging applications, was published in May 2007. The text can be found at:
http://www.dpaonthenet.net/channels/channel1editorial10416.aspx